Physicians frequently make complicated and high-stakes decisions that contradict evidence-based medical guidelines. One psychological reason may be that physicians learn from prior experiences with patients and use their prior experiences when making decisions for future patients. Though learning from prior patients may be adaptive to some extent given that evidence-based guidelines cannot identify optimal decisions for each unique patient, inappropriate reliance on experience may result in suboptimal decisions that fail to adequately incorporate available evidence. Physicians are often expected to make decisions based on guidelines and are expected not to learn from or react to prior patients when treating future patients because experience with prior patients is limited. The overarching goal of this proposal is to elucidate the cognitive and motivational processes through which prior experiences influence physicians' treatment and diagnostic testing decisions. In the first set of studies, physicians will be tracked over time to examine how previous experiences in which patients they discharged and were subsequently readmitted to the hospital influenced the physicians' decisions to discharge future patients from the hospital. This will be accomplished by analyzing an extensive database of medical records for hospitalized patients. Psychological factors such as the level of expertise of the physician, the source of prior experiences (personal experience or social transmission), and how physicians generalize knowledge from patient to patient will be examined for influences on discharge decisions. In a second set of studies examining the effects of experience on decision making, medical students, interns, and residents will engage in a series of simulated decisions about hypothetical patients who may have pulmonary embolism. Testing for pulmonary embolism involves a complicated set of decisions, and lab-based experiments will allow us to carefully examine participants' testing strategies as they change from one patient to the next. Psychological factors such as the prevalence of pulmonary embolism, the order of pulmonary embolism cases, and whether participants learn that specific symptoms predict pulmonary embolism will be examined for influences on testing strategies. Collectively, the proposed studies will elucidate psychological reasons for sub- optimal medical decisions and may suggest ways to improve medical decision-making.